Preventing Aggressionin the Licensed Environment

Ross HomelGriffith Institute for Social and Behavioural Research

NSW Bureau of Crime Statistics and ResearchOccasional Seminar

October 16, 2009

OutlineOutline

The context of the problemThe main factors and processes that lead to violenceWhat works?: developing evidence-based preventive practicesSustainable regulationAn experiment to generate new knowledge

The context of the problemThe main factors and processes that lead to violenceWhat works?: developing evidence-based preventive practicesSustainable regulationAn experiment to generate new knowledge

Kate Graham is the main international expert:

Homel’s perspective (and this presentation) is criminological

and regulatory

Kate Graham is the main international expert:

Homel’s perspective (and this presentation) is criminological

and regulatory

The Licensed EnvironmentThe Licensed Environment

The environment inside licensed drinking establishments The immediate external environment (entrances and exits, adjoining public space)‘Entertainment precincts’ – Kings Cross, Surfers Paradise, Chapel St Melbourne etcExcludes homes – but note growing phenomenon of ‘pre-loading’

The environment inside licensed drinking establishments The immediate external environment (entrances and exits, adjoining public space)‘Entertainment precincts’ – Kings Cross, Surfers Paradise, Chapel St Melbourne etcExcludes homes – but note growing phenomenon of ‘pre-loading’

One punch really can kill, guard saysGeorgia Waters, BN | July 7, 2008 - 5:01PM

One punch really can kill, guard saysGeorgia Waters, BN | July 7, 2008 - 5:01PM

The Brisbane security guard seriously assaulted at a Fortitude Valley nightclub over the weekend says he didn't really believe that one punch could kill until he was attacked.Steve Hendry, 22, sustained a fractured skull after being elbowed by a patron at the Empire Hotel on Brunswick Street in the earlyhours of Sunday morning.Mr Hendry, a security guard at the nightclub, was trying to break up a fight between a group of patrons, and had been restraining a man who had attacked a smaller man.He suffered two seizures and was taken to the Royal Brisbane Hospital, where he spent about eight hours on life support and 30 in intensive care.

The Brisbane security guard seriously assaulted at a Fortitude Valley nightclub over the weekend says he didn't really believe that one punch could kill until he was attacked.Steve Hendry, 22, sustained a fractured skull after being elbowed by a patron at the Empire Hotel on Brunswick Street in the earlyhours of Sunday morning.Mr Hendry, a security guard at the nightclub, was trying to break up a fight between a group of patrons, and had been restraining a man who had attacked a smaller man.He suffered two seizures and was taken to the Royal Brisbane Hospital, where he spent about eight hours on life support and 30 in intensive care.

Societal response(the GFC ‘in miniature’?)Societal response

(the GFC ‘in miniature’?)

Make the individual responsible (e.g. pub bans)The deserved misfortune of victimsTough enforcementConstruct the problem solely in terms of alcoholNot enough attention to the role of situational factors, venue management, social systems, and industry culture and structuresA deregulated environment - reviving city or state economies rather than attending to public health outcomes

Make the individual responsible (e.g. pub bans)The deserved misfortune of victimsTough enforcementConstruct the problem solely in terms of alcoholNot enough attention to the role of situational factors, venue management, social systems, and industry culture and structuresA deregulated environment - reviving city or state economies rather than attending to public health outcomes

The Global Context•Youth framed as a problem: the risk-taking generation•Concept of ‘rights-based’citizenship: the new citizen must be a responsible citizen•Emphasis on constructing oneself, making choices, having one’s own biographical project•Policies focus on getting young people (and parents) into work: building human capital, meeting the needs of the labour market•Not much room for young people to make mistakes•Hold parents responsible - and punish offending young people

Drinking EstablishmentsDrinking Establishments

Alcohol - makes people:More focused on the presentLess aware of internal thought processesBut effects determined by cultural and group expectations

‘Time out’ from daily life : licensed venues do not just sell alcoholThe social functions of public drinking establishments - and the kinds of people who go there - make them a high risk setting for aggression

Alcohol - makes people:More focused on the presentLess aware of internal thought processesBut effects determined by cultural and group expectations

‘Time out’ from daily life : licensed venues do not just sell alcoholThe social functions of public drinking establishments - and the kinds of people who go there - make them a high risk setting for aggression

Statistical evidenceStatistical evidence

A greater density of premises = more violenceSynergistic or non-linear effects (Livingston et al.)

Licensed premises account for between one third and one half of all assaults

Last Drinks surveysSurvey data shows that adults aged 18-30 experience aggression most commonly in bars, clubs etc - more so for serious violenceEmployment in drinking establishments also a risk factor

A greater density of premises = more violenceSynergistic or non-linear effects (Livingston et al.)

Licensed premises account for between one third and one half of all assaults

Last Drinks surveysSurvey data shows that adults aged 18-30 experience aggression most commonly in bars, clubs etc - more so for serious violenceEmployment in drinking establishments also a risk factor

Alcohol and violenceAlcohol and violence

Aggression occurs when there is a combination of:

The pharmacological effects of alcoholA person who is willing to be aggressive when drinkingAn immediate drinking context conducive to aggressionA broader cultural context that is tolerant of alcohol-related aggression

Aggression occurs when there is a combination of:

The pharmacological effects of alcoholA person who is willing to be aggressive when drinkingAn immediate drinking context conducive to aggressionA broader cultural context that is tolerant of alcohol-related aggression

Specific processesSpecific processes

Risk takingFocus on the here and now: hyper-emotional effectsCognitive functioningConcern with personal power: the ‘macho’or ‘masculinised’ culture of pubs and clubsPeople are aware of alcohol’s effects on other people but not on them

Risk takingFocus on the here and now: hyper-emotional effectsCognitive functioningConcern with personal power: the ‘macho’or ‘masculinised’ culture of pubs and clubsPeople are aware of alcohol’s effects on other people but not on them

Patron, staff & environmental factors

Patron, staff & environmental factors

13 empirical studies of the relationship between aggression and aspects of drinking establishments

More than half Canadian (Graham) or Australian (Homel)Mostly observational but some interview studies and one phone survey of young adultsMostly quantitative but some qualitativeDifferent measures used in different studiesProduce correlational data - causal processes need experimental confirmation

13 empirical studies of the relationship between aggression and aspects of drinking establishments

More than half Canadian (Graham) or Australian (Homel)Mostly observational but some interview studies and one phone survey of young adultsMostly quantitative but some qualitativeDifferent measures used in different studiesProduce correlational data - causal processes need experimental confirmation

Patron risk factorsPatron risk factors

Young (but depends on study and specific indicator)Indigenous or other specific ethnic groups‘Marginal’ patronsSalience of some common individual risk factors for criminality (e.g., impulsivity, heavy drinkers)Machismo - patrons and staff (esp. security)

Young (but depends on study and specific indicator)Indigenous or other specific ethnic groups‘Marginal’ patronsSalience of some common individual risk factors for criminality (e.g., impulsivity, heavy drinkers)Machismo - patrons and staff (esp. security)

Physical environmentPhysical environment

Line-ups and people milling around outsideSize of establishmentDirty premisesCrowdingDiscomfort - lack of seating (vertical drinking), smoky air, inconvenient bar access

Line-ups and people milling around outsideSize of establishmentDirty premisesCrowdingDiscomfort - lack of seating (vertical drinking), smoky air, inconvenient bar access

Exit Only

To & From Bar To & From Bar

Entrance & Exit

Entrance & Exit

Entrance & Exit

Entrance & Exit

To & From Bar

To & From Bar

To & From Toilets

To & From Dancefloor

To & From Bar

To & From Bar

LOW RISK NIGHTCLUB

HIGH RISK NIGHTCLUB

0.80.60.40.20.0-100

0

100

200

300

DENSITYPERSONS PER SqM

High Risk Venues

Low Risk Venues

Social environmentSocial environment

Permissive environments in which rules and limits are unclear, especially around dancing and pool playingOverall levels of intoxicationSexual activity, sexual competitionIllegal activities, drug dealing, prostitution

Permissive environments in which rules and limits are unclear, especially around dancing and pool playingOverall levels of intoxicationSexual activity, sexual competitionIllegal activities, drug dealing, prostitution

StaffStaff

Proliferation of premises with specialised and gendered staff roles - especially security staffNumbers may be too few or too manyStaff skills (e.g., monitoring and defusing situations)Lack of responsible serving‘Bad apples’Enforcer versus guardian role

Proliferation of premises with specialised and gendered staff roles - especially security staffNumbers may be too few or too manyStaff skills (e.g., monitoring and defusing situations)Lack of responsible serving‘Bad apples’Enforcer versus guardian role

The external environmentThe external environment

Spilling out the doorsHot spotsGeographically isolated venues can be a bigger problem than ‘entertainment areas’Not all areas with a high density of venues are a problemThe symbolic landscapeNeed much more finely grained, small area data to understand the ecology of licensed environments

Spilling out the doorsHot spotsGeographically isolated venues can be a bigger problem than ‘entertainment areas’Not all areas with a high density of venues are a problemThe symbolic landscapeNeed much more finely grained, small area data to understand the ecology of licensed environments

Governance through partnerships(“It should be a working together thing”)

Governance through partnerships(“It should be a working together thing”)

Many forms of partnerships internationallyAlcohol AccordsCrime and Disorder PartnershipsConcierge governmentCommunity action coalitions

Social control:Local and licensing lawsPolicing strategies: reactive vs problem-solvingCrime prevention through environmental design

Many forms of partnerships internationallyAlcohol AccordsCrime and Disorder PartnershipsConcierge governmentCommunity action coalitions

Social control:Local and licensing lawsPolicing strategies: reactive vs problem-solvingCrime prevention through environmental design

What works to prevent violence?What works to prevent violence?

High quality staff training grounded in research: the Ontario Safer Bars Program (encouraging evidence for long-term effects)Some forms of police and regulatory enforcementof licensing laws and responsible management practices (long-term effects not demonstrated)Community action models (long-term effects only achieved in Sweden)

High quality staff training grounded in research: the Ontario Safer Bars Program (encouraging evidence for long-term effects)Some forms of police and regulatory enforcementof licensing laws and responsible management practices (long-term effects not demonstrated)Community action models (long-term effects only achieved in Sweden)

Safer BarsSafer Bars

Three-hour training program for all staff and management - reducing and managing problem behaviour and aggressionA risk assessment workbook for managersLegal pamphletEvaluated through a large scale randomized controlled trial: 26 large capacity bars and 12 similar controls

Three-hour training program for all staff and management - reducing and managing problem behaviour and aggressionA risk assessment workbook for managersLegal pamphletEvaluated through a large scale randomized controlled trial: 26 large capacity bars and 12 similar controls

Improvements in knowledge & attitudes Improvements in knowledge & attitudes by years of experience in the industryby years of experience in the industry

5 6 7 8 9 10

more than 10 years

5-10 years

2-5 years

1-2 years

one year or less

5 6 7 8 9 10

more than 10 years

5-10 years

2-5 years

1-2 years

one year or less

Average number of incidents per observationAverage number of incidents per observationinvolving severe aggression by patronsinvolving severe aggression by patrons

Control ExperimentalControl Experimental

PolicingPolicing

Value of randomised (Jeffs & Saunders) approach not clearBest experiment has been in Wellington, New Zealand (randomised plus targeted)Combined regulatory approaches may be better -but very limited evidence for the effectiveness of Licensing Accords (partial exception of Cardiff)Targeted/Last drinks approach very promising(Wiggers’ research in NSW and the NSW Police Alcohol Linking Program)

Value of randomised (Jeffs & Saunders) approach not clearBest experiment has been in Wellington, New Zealand (randomised plus targeted)Combined regulatory approaches may be better -but very limited evidence for the effectiveness of Licensing Accords (partial exception of Cardiff)Targeted/Last drinks approach very promising(Wiggers’ research in NSW and the NSW Police Alcohol Linking Program)

The NSW Alcohol Linking Model

The NSW Alcohol Linking Model

Emphasis on sustainability from the outset‘Research into practice’ team formedCritical role of Last Drinks data: established feasibility of long-term police commitmentIntervention involved:

Feedback report to licenseesEducational visits by police to offending establishmentsFollow-up workshop for visited licensees

36% reduction in alcohol-related incidents in experimental group vs controls over 3 months (N=400) - but smaller reduction in assaults

Emphasis on sustainability from the outset‘Research into practice’ team formedCritical role of Last Drinks data: established feasibility of long-term police commitmentIntervention involved:

Feedback report to licenseesEducational visits by police to offending establishmentsFollow-up workshop for visited licensees

36% reduction in alcohol-related incidents in experimental group vs controls over 3 months (N=400) - but smaller reduction in assaults

Wellington experimentWellington experiment

Combined regulatory approachHeightened police presence in premises at night during 2 6-week periods (30 minute visits)Public health & licensing officials made daytime visits - educative/compliance focusAlso targeted 20 problem premises

Combined regulatory approachHeightened police presence in premises at night during 2 6-week periods (30 minute visits)Public health & licensing officials made daytime visits - educative/compliance focusAlso targeted 20 problem premises

Strategies for achieving sustained effects

Strategies for achieving sustained effects

Obtaining organizational leadership and policy support…Providing supportive organizational infrastructure…Developing police knowledge and skills…Implementing data quality assurance and performance feedback strategies.Now being adopted across Australia and in New Zealand

Obtaining organizational leadership and policy support…Providing supportive organizational infrastructure…Developing police knowledge and skills…Implementing data quality assurance and performance feedback strategies.Now being adopted across Australia and in New Zealand

Community ActionQueensland Safety Action Model (1990s)

Community ActionQueensland Safety Action Model (1990s)

Community forum/ community-based task groups plus safety audit;Risk assessments in licensed premises by project personnel, and Code of Practice by nightclub managers; Training of the community-based project steering committee, the project officer, managers, bar and security staff, and police; Improvements in external regulation of licensed premises by police and liquor licensing inspectors

Community forum/ community-based task groups plus safety audit;Risk assessments in licensed premises by project personnel, and Code of Practice by nightclub managers; Training of the community-based project steering committee, the project officer, managers, bar and security staff, and police; Improvements in external regulation of licensed premises by police and liquor licensing inspectors

Level of regulation: responsive regulation model

Level of regulation: responsive regulation model

State: Formal Regulation and Law Enforcement

Local Community: Informal controls and persuasion

Venues: Self-regulation by licensees

Changes in overall numbers ofassaults between 1993 and 1996

0

2

4

6

8

10

12

1993 1994 1996 1999

Surfers ParadiseNorth Queensland

The Stockholm Prevents Alcohol & Drug Problems (STAD) Project

The Stockholm Prevents Alcohol & Drug Problems (STAD) Project

Survey (mid-1990s) of owners of licensed premises: owners saw no problems with overservingFormation of action group: develop strategies to prevent intoxication and service to minors

Two-day training course in RBS for servers, security staff and owners;New forms of enforcement: notification letters, mutual controls (police & licensing officials)

Signing of written agreement by high-ranking officials --> formal steering committee

Survey (mid-1990s) of owners of licensed premises: owners saw no problems with overservingFormation of action group: develop strategies to prevent intoxication and service to minors

Two-day training course in RBS for servers, security staff and owners;New forms of enforcement: notification letters, mutual controls (police & licensing officials)

Signing of written agreement by high-ranking officials --> formal steering committee

STAD outcomesSTAD outcomes

Interrupted time series analyses of police-recorded violence (inside & outside) between 10 pm and 6 am

Reduction of 29% in intervention area, slight increase in control area

Gradual reduction as interventions became more intense

No displacement or influence of extraneous factors

Increase in rates of refusal of service to drunks (5% --> 70%)

Effects sustained over a period of 5 or more years

Interrupted time series analyses of police-recorded violence (inside & outside) between 10 pm and 6 am

Reduction of 29% in intervention area, slight increase in control area

Gradual reduction as interventions became more intense

No displacement or influence of extraneous factors

Increase in rates of refusal of service to drunks (5% --> 70%)

Effects sustained over a period of 5 or more years

STAD sustainabilitySTAD sustainability

Strong inter-agency collaborative climate with strong leadership from head of licensingLobbying by action group membersPolice gradually came on board stronglyInstitutionalisation through agency financial support, signed agreement10 year time framePartnership of agencies: “community” only involved through venue employees, leisure industry etcCurrently being extended to all local government areas in Sweden

Strong inter-agency collaborative climate with strong leadership from head of licensingLobbying by action group membersPolice gradually came on board stronglyInstitutionalisation through agency financial support, signed agreement10 year time framePartnership of agencies: “community” only involved through venue employees, leisure industry etcCurrently being extended to all local government areas in Sweden

ConclusionsConclusions

Using a responsive regulation framework:Develop local partnerships oriented to evidenceUnderstand the local ecology, industry climate and regulatory systemsIncorporate Safer Bars training/risk assessmentsuniversally Experiment with randomised & targeted enforcement using last drinks dataBuild sustainability mechanisms in from the outsetBuild community coalitions /action groups suitable to local conditions

Using a responsive regulation framework:Develop local partnerships oriented to evidenceUnderstand the local ecology, industry climate and regulatory systemsIncorporate Safer Bars training/risk assessmentsuniversally Experiment with randomised & targeted enforcement using last drinks dataBuild sustainability mechanisms in from the outsetBuild community coalitions /action groups suitable to local conditions

Sustaining a reduction of alcohol-related harms in the licensed

environment:A practical experiment

to generate new evidence

Key Centre for Ethics, Law, Justice and GovernanceGriffith University, Brisbane, Queensland

February 2009

Why this Project?Review of international field revealed very little in the way of long term effectiveness (“Raising the Bar”, Graham & Homel, 2008)

Transition from research into practice poorly done – need rigorous research design that can translate to easily implemented practice:– Simple, practical and evidence based

Unable to sustain reductions in our work in Surfers Paradise and northern Qld

Consistent indications of others unable to sustain reductions

In Surfers Paradise – ‘return’ to base line levels of harm

Reduction of 73% in serious assaults achieved with the Safety Action Project (1996) largely lost by 1999

In 2007:• 66% of ambulance calls• 27% of assaults

• 18% sexual assaults……..were alcohol related

The health and injury costs of licensed venues are high

Our challenge……..

Could we develop a comprehensive prevention model that:– Was capable of sustaining reductions in

harm/violence around licensed premises– Could be “fitted” to various jurisdictions– Was constantly reviewed– Remained evidence-based– Could eventually be institutionalised in

communities?

Yes we could!

The Model

Framework: Responsive Regulation

1. “Safer Bars” training for all venue staff

2. Community Mobilisation - Informal regulation

3. Policing - targeted using Last Drinks methods based on ambulance, ED & police data

Two Phase Research Project

Phase 1: 12 months (2008/9) to:– Develop a scientifically defensible research

design – Select and “fit” model at 5 sites in Australia

and New Zealand

Phase 2: Trial the model for 7 years at each site

First Phase

5 sites selected:Mackay - QueenslandWellington - New ZealandInner CBD Melbourne - VictoriaChapel Street - VictoriaSt. Kilda – Victoria

Research Design finalised

Expert Group

Expert groupMr. Neil Comrie

Dr. John Wiggers

Mr. Paul Dillon

Mr Michael Lockwood

Prof Ross Homel

Prof Paul Mazerolle

Second Phase

Evaluate impact of model at each of the 5 sites

Institutionalise model to ensure sustained reductions in alcohol related violence and harm

5 -7 years minimum

What will Griffith Uni do?

Add value to the resources in communities

Collect and analyse data provided by sites and feed back reports for each site to:– Target policing– Develop local strategies– Inform service resourcing– Informal local and wider policy

Conduct the larger meta-experiment evaluating all 5 sites

What is so different?

Combining data components (Emergency Dept., Ambulance and Police)

Integrating injury prevention, public health, community safety and crime prevention

Balancing formal, informal (Monitoring Committee) and self regulation (Licensees)

Long term – mechanisms will be integrated into communities as NORMAL practice

Internationalisation

John Moores Uni – Prof. Mark Bellis:Comparison of KaREN assessment

Karolinska Institute – Prof. Sven Andreasson

PhD student for one year (Mats Halgren)

Centre for Addiction and Mental Health, Canada - Kate Graham :

Comparison of Safer Bars

What is exciting about this research?

The commitment from all 5 sites is highOpportunity to make a difference long term, with international partnershipsOpportunity to leave permanent practices behind after research project has finishedOpportunity to make drinking environments safer and less harmful for young people………………..